FR2812656A1 - NUCLEIC ACIDS AND PROTEINS FROM E. COLI PATHOGENS, AND THEIR USES - Google Patents

Abstract

The invention concerns novel markers of <i>Escherichia coli</i> pathogenicity, useful in particular for detecting pathogenic <i>Escherichia coli</i> of serotype O157 producing Shiga-like toxins.

Description

<Desc / Clms Page number 1>

NUCLEIC ACIDS AND PROTEINS FROM E. COLI PATHOGENS, AND THEIR
USES.

The present invention relates to new pathogenicity markers of Escherichia coli, which can be used in particular for the detection of pathogenic Escherichia coli of serotype 0157 producing shiga toxins.

Escherichia coli (E. coli) is a ubiquitous bacterium, frequently isolated in medical or food microbiology laboratories, and present in the normal state in the intestinal flora of humans. However, some strains are pathogenic, and associated with many intestinal and non-intestinal infections.

The identification of the different strains of E. coli is conventionally based on the serological typing of their 0 (somatic antigen) and H (flagellar antigen) antigens by immunological methods.

The serotypes of E. The most common pathogenic coli have been classified into 5 main groups, according to their mechanism of infection of the digestive tract:
EPEC, for: "enteropathogenic E. coli", which groups together classical enteropathogens, - ETEC, for: "enterotoxigenic E. coli", which groups together E. coli producing a stable or heat-sensitive toxin - EAGGEC, for "enteroagregative E. coli ", which groups together the E. coli which are characterized by a particular adhesion profile called" aggregative adhesion ",
EIEC, for: "enteroinvasive E. coli", which groups together the E. coli which mimic the ability of Shigella to invade and multiply in the cells of the intestinal epithelium and - VTEC, for: "verotoxin producing E. coli" , or STEC "shiga-toxin producing E. coli", which groups together the E. coli producing cytotoxins, called "Shiga-like toxins", because of their analogy with the toxin secreted by Shigella dysenteriae type 1 or also verotoxins ( VT), due to their toxic activity on Vero cells.

<Desc / Clms Page number 2>

Among the strains of E. pathogenic coli, the colibacilli producing shiga-toxins called STEC, are emerging microorganisms increasingly recognized as responsible for food poisoning (GANNON et al., 1993, J. Clin. Microbiol. 31: 1268-1274; BEGUM D. et al., 1993, J. Clin. Microbiol. 31: (12) 3153-3156). In humans, STEC infection results in more or less severe symptoms which can range from simple diarrhea, hemorrhagic colitis, to hemolytic uremic syndrome (HUS) or thrombotic thrombocytopenic purpura ( PTT), and can sometimes be fatal (GRIFFIN et al., 1991, Epidemiol. Rev. 13: 60-98; NATARO et al., 1998, Microbiol.

Immunol. 42: 371-376; PATON et al., 1998, Clin. Microbiol.

Rev. 11: 450-479).

Hemolytic uremic syndrome occurs mainly in young children; it accounts for about 9% of STEC infections diagnosed in the United States. It is also the main cause of acute renal failure in children, and the lethality observed is of the order of 3 to 5/1000. In France, recent studies have shown that the average annual incidence of HUS is 0.7 / 10,000 in children under 15 and 1.8 / 10,000 in children under 5 (DECLUDT et al. ., 2000, Epidemiol. Infect. 124: 215-220). STECs are also involved in animals, in more or less serious digestive infections. Symptoms observed include diarrhea and dysentery in calves or edema disease in pigs. In addition, cattle, healthy carriers, could play a role in the transmission of STECs to humans, through contaminated water or foodstuffs of animal origin (meat, raw milk and dairy products based on raw milk), (BONNET et al., 1998, J. Clin. Microbiol. 36: 1777-1780; PRADEL et al., 2000, J. Clin. Microbiol. 38: 1023-1031).

Initially, only one serotype of E. coli had been associated with HUS: this is serotype 0157: H7. It was subsequently observed that other serotypes including 0157: H-, 026: H11, 0111: H- and 0103: H2 could be involved.

<Desc / Clms Page number 3>

in STEC infections (BRIAN et al., 1992, J. Clin. Microbiol. 30: 1801-1806). Among these serotypes, 0157 is the major serotype, involved in 75 to 80% of hemolytic and uremic syndromes (DECLUDT et al., Cited above).

The shiga-toxins produced by STECs, in particular SLT1 (VT1), SLT2 (VT2) and their variants VT2 vha, VT2 vhb, SLTII-v (VTE) and SLT II-va (VTE-va) (ZAWLIN et al., 1993, Microbiol. Immunol., 37: 543-548), have been identified as the main pathogenicity factors involved in these infections (CALDERWOOD et al., 1987, Proc. Nat. Acad.

Sci. 84: 4364-4368).

However, shiga toxins alone are not responsible for the pathogenicity of STECs; other bacterial virulence factors could also play a role in the pathogenicity of these colibacilli: - the enterocyte erasure locus (LEE) which codes for proteins involved in the adhesion of bacteria and the alteration of the cytoskeleton of intestinal cells , in particular the intimin encoded by the eae gene (BEEBAKHEE et al., 1992, FEMS Microbiol. Lett. 70: 63-68; Mc DANIEL et al., 1995, Proc. Nat. Acad., Sci. 92: 1664- 1668; Mc DANIEL et al., 1997, Mol. Microbiol. 23: 399-407), - the efa 1 gene (EHEC factor for adherence) has been described as a factor which contributes to the adhesion capacity of the bacterium ( NICHOLLS et al., 2000, Mol.

Microbiol., 35: 275-288), and - a 90 kb plasmid (BURLAND et al., 1998, Nucleic Acids Res., 26: 4196-4204), which encodes potential virulence factors such as enterohemolysin ( SCHMIDT et al., 1995, Infect. Immun., 63: 1055-1061; SCHMIDT et al., 1996, Microbiol., 142: 907-914)), a catalase-peroxidase (BRUNDER et al., 1998, Microbiol. , 142: 3305-3315), a serine protease (BRUNDER et al., 1997, Mol. Mirobiol., 24, 767-778) and a type II secretion system (SCHMIDT et al., 1997, FEMS Microbiol. Letter, 148: 265-272).

To prevent pathologies associated with shiga-toxin-producing colibacilli (STEC) and in

<Desc / Clms Page number 4>

particularly the most serious pathologies (hemorrhagic colitis, HUS, PTT) associated with the most pathogenic among them, it is necessary to have means of detecting these colibacilli, in order to be able to systematically control the water and foods liable to be contaminated, and in particular fresh products of animal origin (meat, raw milk and dairy products made from raw milk).

This control involves the sensitive and specific detection of STECs responsible for these pathologies and in particular of STEC serotype 0157.

This detection is complex because, as mentioned above, the pathogenicity of STEC is not simply linked to a particular serotype, but that it results from the association of a certain number of virulence factors, the combination of which exact is not known. In addition, certain virulence factors are common to STECs and to other groups of pathogenic E. coli or to other species of bacteria, which constitutes an additional problem for the specific detection of STECs and in particular of STECs of serotype 0157. For example, the 90 kb virulence plasmid is present in pathogenic STECs of other serotypes (BONNET et al., supra), and the LEE locus is present in both pathogenic STECs of different serotypes (PRADEL et al., cited above) and in EPECs, in particular in 0127: H6 (PERNA et al., 1998, Infect. Immun., 66: 3810-3817; PRADEL et al., cited above).

Cytotoxicity tests and various biochemical, immunological (ELISA) or genetic methods have been proposed to detect STEC (NATARO et al., 1998, Clin.

Microbiol. Rev., 11: 142-201; PATON et al., 1998, Clin.

Microbiol. Rev., 11: 450-479; FENG et al., 1993, Mol. Cell.

Probes, 7: 151-154; HUCK et al., Int. J. Food Microbiol., 25: 277-287).

# Cytotoxicity tests on cell cultures constitute the reference method used in food hygiene to detect the presence of bacteria producing shiga-toxins or verotoxins. The principle of

<Desc / Clms Page number 5>

these tests are based on the cytopathogenicity of verotoxins on African green monkey kidney cells. Subsequently, confirmation of the antigenic types of verotoxin is required using anti-VT1 and anti-VT2 antibodies.

This technique has the advantage of being very sensitive, however it is not specific for colibacilli producing shiga-toxins of serotype 0157. In addition, a series of transfers in selective media is necessary in order to promote the development of verotoxic bacteria. prior to the assay of toxins on a cell culture; it takes 5-6 days to get the first results, and the costs are high. This test therefore does not constitute a routine method that can be used in the food industry.

# The biochemical methods use the characteristics of the majority of E. coli 0157: H7 and 0157: H-, which do not ferment sorbitol in 24 hours, and which furthermore do not produce P-glucuronidase. E. coli 0157: H7 and 0157: H 'appear as white colonies on sorbitol agars, while coli-producing coli-producing shiga-toxins of other serotypes (026,0111 and 0103) and the numerous nonpathogenic coli bacteria, which are capable of fermenting sorbitol, appear as pigmented colonies. However, these methods are insufficient because there are atypical mutants of E. coli 0157: H7 and 0157: H- which are capable of fermenting sorbitol (GUNZER et al., 1992, J. Clin.

Microbiol., 30: 1807-1810; KARCH et al., 1997, J. Food, Prot., 60: 1454-1457).

# ELISA-type immunological methods are based on the detection of 0 and H antigens or of shigatoxins (VT1 and VT2) using specific antibodies. The immunological tests currently available are not satisfactory; in fact, other bacteria such as Citrobacter freundii, Salmonella 0 of group N and Hafnia alvei exhibit cross-reactivity with respect to the anti-0157 sera used (BETTELHEIM et al., 1993, J. Clin.

Microbiol., 31: 760-761; BORCZYK et al., 1987, Int. J. Food

<Desc / Clms Page number 6>

Microbiol., 4: 347-349; CHART et al., 1992, Epidemiol.

Infect., 108: 77-85; SHIMADA et al., 1992, Curr.

Microbiol., 25: 215-217).

In addition, most of the available immunological tests are limited to the detection of the 0157 antigen and do not make it possible to differentiate STEC 0157 from non-pathogenic E. coli of serotype 0157. Finally, these methods are generally too insensitive or too sensitive. not very specific to allow detection in food.

# Polymerase chain reaction (PCR) methods, allowing the specific detection of genes encoding bacterial factors involved in virulence have been proposed. The amplification of the following genes has been reported: - the genes for shiga-toxins (stx), (CHEN et al., 1998, Appl. Environ. Microbiol., 64: 147-152; Application EP 0 864 657 in the name of CNEVA, GANNON et al., 1992, Appl. Environ.

Microbiol., 58: 3809-3818; JACKSON, 1991; J. Clin.

Microbiol., 29: 1910-1914; KARCH et al., 1989, J. Clin.

Microbiol., 27: 2751-2757; LIN et al., 1993, Microbiol.

Immunol., 37: 543-548; POLLARD et al., J. Clin. Microbiol., 1990, 28: 540-545; POLLARD, 1990, J. Infect. Dis., 162, 1195-1198; READ et al., 1992, Mol. Cell. Probes, 6: 153-161), - the intimin (eae) gene carried by the LEE locus (SANDHU et al., 1996, Epidemiol. Infect., 116: 1-7; WILLSHAW et al., 1994, J. Clin. Microbiol., 32: 897-902), - the hemolysin gene encoded by the 90 kb plasmid (E-hlya), (SCHMIDT et al., 1995, cited above), - genes involved in the biosynthesis of the 0 antigen (rfb locus), (DESMARCHELIER et al., 1998, J.

Clin. Microbiol., 36: 1801-1804; MAURER et al., 1999, Appl.

About. Microbiol., 65: 2954-2960) and - the gene encoding the H7 flagellar antigen (fliCh7), (FIELDS et al., J. Clin. Microbiol., 1997, 35: 1066-1070).

These methods are more sensitive and more specific than biochemical or immunological methods.

<Desc / Clms Page number 7>

However, STEC 0157 cannot be identified using a single PCR reaction targeting one of these genes. Indeed, the somatic 0157 and flagellar H7 antigens, which can be carried by strains of E. non-pathogenic coli alone are not markers of pathogenicity. In addition, verotoxin genes are by definition carried by all STEC strains; the virulence factors described on the 90 kb plasmid (enterohemolysin, catalase peroxidase, serine protease) are present in STECs of different serotypes (BONNET et al., cited above) and a homolog of the LEE locus is present both in STEC of different serotypes (PRADEL et al., Cited above) and in EPECs (PERNA et al., Cited above). Consequently, the demonstration of only one of these virulence genes is not sufficient to identify a STEC of serotype 0157.

Tests allowing the simultaneous amplification of several genes in a single reaction (PCR-multiplex) have been proposed (CEBULA et al., 1995, J. Clin. Microbiol., 33: 1048; DENG et al., 1996, J. Food Prot., 59: 570-576; FRATAMICO et al., 1995, J. Clin. Microbiol., 33: 2188-2191; GANNON et al., 1997, J. Clin. Microbiol., 35: 656-662; NAGANO et al., 1998, Microbiol. Immunol., 42: 371-376; PATON et al., 1998, J. Clin. Microbiol., 36: 598-602).

These techniques, which are suitable for the detection of an isolated bacterial strain, cannot be used for the analysis of complex biological samples, such as food or stool, which are liable to contain a mixture of bacteria. Indeed, as reported by DENG et al. (cited above), PCR-multiplex does not make it possible to conclude whether the signals corresponding to the different genes detected come from a single bacterial strain, or else from the additive effect of genes present in different strains of bacteria. This major drawback, which involves the systematic isolation of each strain in order to verify that the different genes amplified by PCR are indeed present in the same strain, is not suitable for diagnosis.

<Desc / Clms Page number 8>

routine in food hygiene laboratory or medical analysis.

Comparison of the nucleotide sequence of certain genes common to STEC serotype 0157 and other pathogenic E. coli has made it possible to identify point mutations. Thus, certain authors have proposed specific primers targeting the genes carried by STEC 0157 which allow their detection in a single PCR reaction. Thus, the amplification of the following regions has been reported: the 3 'region of the eae gene (LOUIE et al., 1994, Epidemiol. Infect., 112: 449-461), the region immediately adjacent to the 5' region of the gene eae (MENG et al., 1996, Int.

J. Food Microbiol., 32, 103-113; MENG et al., 1997, Lett.

Appl. Microbiol., 24: 172-176; ZHAO et al., 1995, FEMS Microbiol. Lett., 133: 35-39) or the H7 gene (WANG et al., 2000, J. Clin. Microbiol., 38: 1786-1790).

It therefore appears that at the present time there are very few specific markers which are easy to use for detecting pathogenic E. coli, and in particular STEC serotype 0157.

The inventors have identified strains of E. pathogenic coli, in particular in all pathogenic coli bacteria of serotype 0157 producing shiga-toxins, the presence of a nucleic acid insert representing a new locus, absent in non-pathogenic strains. This locus is inserted between positions 93848 and 93849 of the sequence of the genome of E. coli K12 (GenBank U82664); insertion site is precisely located between the stop codons of two open reading frames (Genbank AAB40241 and Genbank AAB40242) in opposite orientation, which in the genomic sequence of strains of E. non-pathogenic coli such as the type K12 strain are directly adjacent.

The inventors also observed the presence of a similar locus, inserted at the same position, in pathogenic E. coli strains, belonging to serotypes other than 0157. On the other hand, they did not observe any insertion at this position. in all the nonpathogenic strains that were tested.

<Desc / Clms Page number 9>

The presence of an insert at this position therefore constitutes a new genetic marker of virulence useful for the detection of pathogenic strains of E. coli.

The nucleic acid sequences derived from such an insert also constitute useful markers for the detection of pathogenic E. coli strains, and in particular for the specific detection of the pathogenic bacterial strains carrying said insert; example, nucleic acid sequences derived from the insert present in a strain of serotype 0157 producing shiga-toxins allow specific detection of all the E. coli of serotype 0157 producing shiga-toxins.

Consequently, the present invention relates to a nucleic acid molecule derived from a pathogenic strain of E. coli, characterized in that it is selected from the group consisting of: a) a nucleic acid molecule representing a locus inserted into the genome of said pathogenic strain, at the location of said genome corresponding to that located in E.coli K12, between positions 93848 and 93849 of the GenBank sequence U82664; b) a nucleic acid molecule comprising the molecule a) defined above, and between 1 and 2500 bp of the sequence upstream of the site of insertion of said molecule, and / or between 1 and 2500 bp downstream of the site insertion of said molecule. c) a nucleic acid molecule constituting a fragment of at least 15 bp, preferably at least 25 to 30 bp, of the molecule a) above; d) a nucleic acid molecule constituting a fragment of at least 15 bp, preferably at least 25 to 30 bp, of the above molecule b), said fragment comprising at least a portion of the molecule a) and at least a portion of the sequence upstream of the insertion site of said molecule a), or at least a portion of the sequence downstream of the insertion site of said molecule a).

Nucleic acid molecules in accordance with the invention can in particular be derived from the genome

<Desc / Clms Page number 10>

a pathogenic bacterium of the species Escherichia coli, chosen from E.coli STEC serotype 0157,077, 079, or 0145, Escherichia coli STEC or EPEC serotype 055, and Escherichia coli EPEC serotype 0127.

For example, nucleic acid molecules in accordance with the invention derived from the genome of E. coli 0157: H7, are represented in the attached sequence listing under the numbers SEQ ID NO: 1, SEQ ID NO: 2 respectively. , SEQ ID NO: 11 to 17, SEQ 19, SEQ ID No: 21 and SEQ ID NO: 23.

A subject of the present invention is also nucleic acid molecules capable of hybridizing specifically, under stringent conditions, with any one of the molecules a) to d) above.

This includes in particular oligonucleotides which can be used as amplification primers for obtaining a nucleic acid sequence in accordance with the invention, or as probes for the detection of said sequence, and in particular the oligonucleotides P388L, P388M, P388G, P388D, P388F, P388C and P388P, corresponding respectively to the sequences SEQ ID NO: 3 to 9 in the attached sequence listing. The oligonucleotides P388M, P388G, P388D, P388F and P388C can in particular be used as molecular hybridization probes, useful for the detection, of a nucleic acid molecule a) in accordance with the invention derived from the genome of a bacterium E coli STEC serotype 0157.

The present invention encompasses in particular the pairs of primers which can be used for the amplification of a nucleic acid molecule a) b) c) or d) in accordance with the invention.

A subject of the present invention is also the recombinant vectors, and in particular the recombinant plasmids containing a nucleic acid molecule in accordance with the invention.

Nucleic acid molecules in accordance with the invention can, for example, be obtained by polymerase chain reaction, from the DNA of a pathogenic E. coli strain comprising the locus defined above. Especially ':

<Desc / Clms Page number 11>

- a nucleic acid molecule a) in accordance with the invention can be obtained by using a pair of amplification primers, one of which hybridizes comprising a primer hybridizing at each of the ends of the molecule a) - a nucleic acid molecule b) in accordance with the invention can be obtained by using a pair of amplification primers comprising a primer hybridizing upstream of the insertion site of molecule a) and a primer hybridizing downstream of the insertion site of molecule a); appropriate primers can be chosen by those skilled in the art, respectively from the sequence of 2500 bp upstream of position 93848 of the GenBank U82664 sequence, and of the sequence of 2500 bp downstream of position 93849.

Amplification conditions will of course be used which make it possible to amplify fragments of large size, and comprising in particular a step of extending primers of sufficient duration.

Advantageously, one can use the pair of primers P388L / P388P, which hybridizes on either side of the site of insertion of a molecule a) in accordance with the invention.

This pair amplifies a fragment of 682 bp when the amplification is carried out from the genome of the non-pathogenic E.coli type strain K12, while it amplifies a fragment of larger size with pathogenic E.coli; it amplifies in particular a 3316 bp fragment (SEQ ID NO: 11) with all the E.coli producing shiga toxins of serotype 0157.

A nucleic acid molecule c) in accordance with the invention can be obtained by using a pair of primers hybridizing with internal sequences of the molecule a)
A nucleic acid molecule d) in accordance with the invention can be obtained by using a pair of primers comprising a primer hybridizing with an internal sequence of molecule a) and a primer hybridizing with a sequence located upstream of the insertion site of molecule a) or a primer hybridizing with a sequence located downstream of the insertion site of molecule a).

<Desc / Clms Page number 12>

- One can advantageously use a pair of primers selected from the group consisting of: P388C / P388D (SEQ ID NO: 8 / SEQ ID NO: 6), P388F / P388G (SEQ ID NO: 7 / SEQ ID NO: 5) , P388F / P388D (SEQ ID NO: 7 / SEQ ID NO: 6), P388P / P388L (SEQ ID NO: 9 / SEQ ID NO: 3), P388L / P388M (SEQ IDNO: 3 / SEQ ID NO: 4) and P388P / P388D (SEQ ID NO: 9 / SEQ ID NO: 6).

A subject of the present invention is also the use of a nucleic acid molecule in accordance with the invention for the screening of pathogenic E. coli, and in particular for the specific screening of strains of E. coli serotype 0157 producing shiga toxins.

In particular, the present invention relates to a method for detecting pathogenic E. coli, characterized in that it comprises the detection of the presence in the genome of E. coli bacteria present in the sample to be tested, of a locus inserted at the location of said genome corresponding to that located in E.coli K12, between positions 93848 and 93849 of the GenBank U82664 sequence.

According to a preferred embodiment of the detection method in accordance with the invention, it comprises at least one step of polymerase chain reaction using a pair of primers comprising a primer hybridizing with the region of the E.coli genome defined by the sequence of 2500 bp upstream of position 93848 of the sequence GenBank U82664, and a primer hybridizing with the region of the E.coli genome defined by the sequence of 2500 bp downstream of position 93849 of the GenBank U82664 sequence.

The determination of the size of the amplification product, and its comparison with that obtained using the same pair of primers, from the genome of a non-pathogenic strain, for example the type strain E.coli K12, makes it possible to detect the insertion of a locus in accordance with the invention.

In addition, the size of the amplification product for the same pair of primers can also make it possible to distinguish different groups of E. pathogenic coli.

<Desc / Clms Page number 13>

For example, if the pair of primers P388L / P388P is used, it is observed in the presence of E. pathogenic coli of serotype 0157 producing shiga toxins, a 3316 bp amplification fragment specific for these bacteria.

According to one variant, the amplification is carried out under conditions which do not make it possible to amplify large fragments. The absence of amplification product makes it possible, in this case, to detect the insertion of a locus in accordance with the invention. This mode of implementation does not, however, make it possible to distinguish different groups of E. pathogenic coli.

According to one embodiment of the detection method in accordance with the invention, it comprises at least one step of polymerase chain reaction using a pair of primers hybridizing with internal sequences of 'a molecule a) as defined above.

For example, for the detection of STEC of serotype 0157, the use of the following pairs of primers: P388C / P388D, P388F / P388D, P388F / P388G produces amplification fragments of respectively 300 bp (SEQ ID NO: 12), 125 bp (SEQ ID NO: 13) and 578 bp (SEQ ID NO: 14) with all the STECs of serotype 0157 while no amplification product is observed with the type strain of E. coli K12 used as control.

According to yet another one embodiment of the detection method in accordance with the invention, it comprises at least one step of polymerase chain reaction using a pair of primers comprising a primer s' hybridizing with an internal sequence of a molecule a) as defined above, and a primer hybridizing with the region of the E. coli genome defined by the sequence of 2500 bp upstream of position 93848 of the sequence GenBank U82664, or a primer hybridizing with the region of the E. coli genome defined by the sequence 2500 bp downstream of position 93849 of the GenBank U82664 sequence.

For example, for the detection of STEC serotype 0157, the use of the following pairs of primers: P388P / P388D, P388L / P388M, produces fragments of

<Desc / Clms Page number 14>

respectively 1299 bp (SEQ ID NO: 15) and 419 bp (SEQ ID NO: 16) with all the STECs of serotype 0157 whereas no amplification product is observed with the non-pathogenic E. coli strains.

The inventors have studied the sensitivity and specificity of the pairs of primers P388C / P388D, P388F / P388D and P388F / P388G for the detection of STEC-0157.

Out of 211 bacterial strains tested, the three pairs of primers have equivalent sensitivity and detect the 55 STEC-0157 tested as well as the E. coli of serotype 055, which is explained by the clonal origin of these two serotypes (FENG et al., 1998, J. Infect. Dis., 177: 1750-1753; WHITTAM et al., 1993, Infect. Immun., 61: 1619-1629). In addition, with the three pairs of primers, no signal is obtained with the bacteria Citrobacter and Hafnia alvei which exhibit immunological cross reactions with the 0157 antigen.

For the specific detection of all E. coli producing serotype 0157 shiga toxins, the P388F / P388D pair is preferably used.

In fact, this pair of primers has a particularly high specificity and does not give any positive reaction with STECs of other serotypes (30 serotypes tested) as well as with non-STEC E.coli of serotype 0157 or other serotypes ( 55 tested) and bacteria from other species.

The oligonucleotides in accordance with the invention can also be used as primers in a technique of PCR-ELISA type. In this case, it is possible, for example, to use the oligonucleotide P388D previously immobilized on a solid support, as a capture probe to fix the amplification product of 578 bp obtained with the primers P388G and P388F; this amplification product can then be revealed either directly (for example if it has been labeled appropriately by incorporation of labeled nucleotides during the amplification reaction) or by means of a labeled detection probe. '

<Desc / Clms Page number 15>

Different methods for fixing nucleic acids on a solid support, as well as those for labeling them using cold or radioactive labels are well known to those skilled in the art, by way of example, those described in Current Protocols in Molecular will be cited. Biology (Frederick M. AUSUBEL, 2000, Wiley and son Inc, Library of Congress, USA).

A subject of the present invention is also a nucleic acid molecule obtained by polymerase chain reaction of a non-pathogenic E. coli using the pair of primers P388P / P388L. This molecule can be used as a control, in the context of a detection method in accordance with the invention.

Advantageously, said molecule is obtained from E. coli K12 and consists of a fragment of 682 bp (SEQ ID NO: 10).

A subject of the present invention is also any protein encoded by an open reading frame present on a nucleic acid molecule in accordance with the invention, and in particular on a molecule derived from a strain of E. coli of serotype 0157 producing shiga-toxins.

Indeed, from a genomic library of E. coli 0157: H7, the inventors isolated a sequence of 5478 bp (SEQ ID NO: 1) containing a locus in accordance with the invention called SILO, 57, defined by the sequence presented in the list of sequences in the appendix under the number SEQ ID NO: 2. FIG. 1 illustrates the genetic organization of the locus of E. coli 0157: H7, called SIL0157, inserted into the genome of E. coli between positions 93849 and 93848, (with reference to the typical genome of non-pathogenic E. coli K12, Genbank U82664). The regions indicated in white represent the regions homologous between the sequence of non-pathogenic E. coli K12 (Genbank U82664) and the sequence of STEC 0157: H7. The regions indicated in black represent the SILO, 57 locus. The large black arrows comprising accession numbers represent the open reading frames described in non-pathogenic E. coli. The large white arrows represent the potential open reading frames included in the

<Desc / Clms Page number 16>

SILO locus, 57. The small arrows indicate the positions of the different primers used for the PCR amplifications.

The Inventors analyzed this SILO157 locus, and thus demonstrated open reading frames and, in particular 4 open reading frames encoding proteins called hereinafter IHP1, IHP2, IHP3 and IHP4, of 338,140, 120, respectively. and 159 amino acids.

The amino acid sequences of the proteins IHP1, IHP2, IHP3 and IHP4 are presented in the list in the annex, respectively under the numbers SEQ ID NO: 18, 20, 22 and 24, the corresponding nucleotide sequences are presented in the list in the annex, respectively under the numbers SEQ ID NO: 17, 19, 21 and 23.

The proteins in accordance with the invention represent new antigens potentially usable for the detection of pathogenic E. coli.

In particular, in the case of proteins whose genes are carried by the SILO157 locus, mentioned above, the IHP1 protein, which has homologies with proteins of the outer membrane and adhesins of other bacteria (Salmonella flagellin and adhesin of Neisseria, Yersinia, Haemophilus and Staphylococcus) probably constitutes a potentially useful surface antigen for the detection of E. coli producing serotype 0157 shigatoxins.

A subject of the present invention is also any peptide representing a fragment of at least 5 amino acids, and preferably of at least 7 to 15 amino acids, of a protein in accordance with the invention.

A subject of the present invention is also: expression cassettes comprising the nucleotide sequences encoding the proteins or peptides in accordance with the invention, placed under the transcriptional control of an appropriate promoter, in particular a promoter allowing the expression of said proteins or said peptides in transformed host cells.

Advantageously, the expression cassettes contain the

<Desc / Clms Page number 17>

sequences encoding one of the proteins IHP1 to IHP4, as defined above (SEQ ID NO: 17, 19, 21 and 23).

- Recombinant vectors, comprising an insert consisting of the nucleotide sequences encoding the proteins or peptides in accordance with the invention. Advantageously, these vectors are expression vectors comprising at least one expression cassette as defined above.

Numerous vectors where one can insert a nucleic acid molecule of interest in order to introduce it and maintain it in a eukaryotic or prokaryotic host cell, are known in themselves; the choice of an appropriate vector depends on the use envisaged for this vector (for example replication of the sequence of interest, expression of this sequence, maintenance of the sequence in extrachromosomal form or else integration into the chromosomal material of the host ), as well as the nature of the host cell.

A subject of the invention is also prokaryotic or eukaryotic cells transformed with a plasmid or a recombinant vector in accordance with the invention.

The recombinant vectors and the transformed cells in accordance with the invention are useful in particular for the production of proteins and peptides in accordance with the invention.

A subject of the present invention is also antibodies, polyclonal or monoclonal, directed against a protein or a peptide in accordance with the invention.

These antibodies can be obtained by conventional methods, known per se, comprising in particular the immunization of an animal with a protein or a peptide in accordance with the invention, in order to cause it to produce antibodies directed against said protein or said. peptide.

Recombinant protein production and purification techniques, immunization and antibody production methods and immunological techniques based on the detection of antigen-antibody complexes

<Desc / Clms Page number 18>

(ELISA, RIA, Western-Blot) are known in themselves; by way of example, mention may be made of those described in Current Protocols in Molecular Biology (Frederick M. A USUBEL, 2000, Wiley and his Inc, Library of Congress, USA) and in Current protocols in Immunology (John E. Coligan, 2000, Wiley and son Inc, Library of Congress, USA).

A subject of the present invention is also the use of at least one protein, a peptide, or an antibody in accordance with the invention for obtaining a medicament, intended for the prevention or treatment of an infection induced by a pathogenic E. coli bacterium, in particular an E. coli bacterium of serotype 0157 which produces shiga toxins.

A subject of the present invention is also the use of at least one protein, a peptide, or an antibody in accordance with the invention for the immunological detection of pathogenic E. coli, in particular for the screening of producing E. coli serotype 0157. shigatoxins.

The nucleic acid molecules, proteins and peptides, as well as the antibodies in accordance with the invention can also constitute reagents which can be used for the detection of pathogenic strains of E. coli, in particular of all strains of E. coli. producing shiga toxins serotype 0157.

The reagents in accordance with the invention can be used for the search for pathogenic E. coli strains, in particular all E. coli strains of serotype 0157 producing shiga-toxins in food products, in particular water, fresh food products such as meats, raw milk and derived products (cheese, dairy products).

These reagents can also be used for the diagnosis of pathogenic E. coli infections, in particular infections with E. coli producing shiga toxins serotype 0157 of humans and animals, from a biological sample, for example in faeces.

<Desc / Clms Page number 19>

A further subject of the invention is a kit for detecting pathogenic E. coli, in particular all strains of E. coli producing shiga toxins of serotype 0157, characterized in that it includes at least one reagent according to invention.

The present invention will be better understood with the aid of the additional description which follows, which refers to non-limiting examples illustrating the isolation and characterization of a nucleic acid molecule in accordance with the invention, and its use for the identification of pathogenic Escherichia coli, and in particular of Escherichia coli serotype 0157 producing shiga-toxins.

EXAMPLE 1: IDENTIFICATION AND CHARACTERIZATION OF A LOCUS INSERTED IN THE CHROMOSOME OF E. COLI 0157: H7 BY ANALYSIS OF POLYMORPHISM IN DNA BY RAPD (RANDOMLY AMPLIFIED POLYMORPHIC DNA).

1. Identification of a DNA fragment specific for E. coli 0157: H7 by RAPD 1. 1. Extraction of bacterial DNA and PCR amplification
12 strains of E. coli: 10 strains of STEC [3 strains 0157: H7 and 1 strain (0157: H-, 026: H11, 091: H21, 0103: H2, 0111: H +, 0113: H4 and 0128: H2)] and 2 strains of 'E. coli serotype 0157, non-producing shiga-toxins (0157-non STEC) are cultured in TSB medium (Tryptone Soy Broth), at 37 C, overnight and the total bacterial DNA is extracted using the INSTAGENE matrix (BIO-RAD), following the manufacturer's instructions. Strains of E. coli are analyzed by RAPD, using the READY TO GO kit (PHARMACIA BIOTECH). The PCR reaction is carried out in a final volume of 25 l, in the presence of 10 ng of purified total bacterial DNA and 25 pmoles of the oligonucleotide primer P1 (PHARMACIA-BIOTECH). PCR amplification (GENE-AMP PCR SYSTEM 9600, PERKIN ELMER) is carried out under the following conditions: an initial denaturation step at 95 C for 5 min is followed by 45 cycles alternating with a denaturation step at 95 C for 1 min, a hybridization step

<Desc / Clms Page number 20>

primers at 36 C, for 1 min and an extension step at 72 C, for 2 min. The amplification product is separated by electrophoresis on a 2% agarose gel containing ethidium bromide, in TBE buffer (Tris-Borate-EDTA).

The amplified DNA fragments are visualized under an ultraviolet lamp.

The profile of the RAPD analysis shows that a 387 bp fragment is amplified at random, only in the colibacilli producing shiga-toxins of serotype 0157 (3 strains of serotype 0157: H7 and 1 strain of serotype O157: H- ).

1. 2. Hybridization of the fragment obtained by RAPD
The 387 bp fragment is isolated from the agarose gel, purified using the QIAEX II extraction kit (QIAGEN) and cloned into the pMOS Blue vector (AMERSHAM PHARMACIA BIOTECH), using the kit of cloning of blunt ends of AMERSHAM LIFE SCIENCE, to give the plasmid pSP7. A probe corresponding to the 387 bp insert (probe P7) is synthesized by PCR and labeled with digoxigenin using the PCR reagent DIGOXIGENIN LABELING MIX (ROCHE MOLECULAR).

This probe is hybridized, by Dot-blotting, with the DNA of the 12 strains of E. coli analyzed by RAPD. Hybridization is carried out on nylon membranes (ROCHE MOLECULAR), overnight at 65 ° C., in a standard hybridization buffer (ROCHE MOLECULAR) containing 20 ng / ml of denatured probe labeled with digoxigenin, following the protocols classics described in MANIATIS et al. (Molecular Cloning: a laboratory manual, 1982, Cold Spring Harbor Laboartory, Cold Spring Harbor, New York). Then, the hybridization of the probe labeled with digoxigenin is detected by luminescence (DIG LUMINESCENT DETECTION KIT, ROCHE MOLECULAR) on BIOMAX films (KODAK).

The hybridization results show that the labeled P7 probe recognizes only STEC serotype 0157.

This result suggests that the 387 bp fragment obtained by RAPD could originate from a locus specific to the strains of colibacilli producing shiga toxins of serotype 0157.

<Desc / Clms Page number 21>

2. Identification and location of a locus inserted in the chromosome of E. coli 0157: H7.

2. 1 Construction of a DNA library of E. coli 0157: H7
The chromosomal DNA of strain EDL 933 (ATCC 43895) of E. coli 0157: H7, purified using the QIAGEN GENOMIC TIP SYSTEM kit (QIAGEN) is digested, in its entirety, with Hind III (ROCHE MOLECULAR), following the manufacturer's instructions. Then 300 ng of this digested DNA is ligated with 10 ng of the plasmid pUC 18, digested with Hind III and dephosphorylated (PHARMACIA-BIOTECH), in the presence of 5 IU of T4 DNA ligase (ROCHE MOLECULAR), in a volume 20 µl reaction at 16 ° C overnight. Competent E. coli MOS Blue bacteria are transformed with the ligation product and spread on LB-agar dishes containing 100 g / ml of ampicillin, 25 g / ml of IPTG isopropyl PD-thiogalactopyranoside and 25 g / ml of X-gal (5-bromo-4-chloro-3-indolylD-galactoside) and incubated overnight at 37 ° C., according to the standard protocols described in MANIATIS et al., cited above.

The presence of the fragment isolated by RAPD in the plasmid DNA of colonies resistant to ampicillin is analyzed by homologous hybridization using the P7 probe.

The analysis of 1000 colonies of transformed bacteria made it possible to isolate the plasmid pSP26 containing a 4.4 kb fragment of the genome of E. coli 0157: H7.

2. 2. Location of the locus inserted in the chromosome of E. coli 0157: H7
The plasmids pSP7 and pSP26 were sequenced by standard techniques and the sequence obtained was compared with the sequences available in the databases using the BLAST program.

The alignment of the sequences of the plasmids pSP7 and pSP26 shows that the sequence of the 387 bp insert of the plasmid pSP7 comprises the 3 'end of the sequence of the 4.4 kb insert of the plasmid pSP26, to which 127 nucleotides have been added. additional, located at its 3 'end (Figure 1).

<Desc / Clms Page number 22>

Analysis of the sequence of the 4.4 kb fragment of pSP26 shows that its 5 'end has 98% identity with the genomic sequence of minutes 9 to 11 (positions 96304-93849, with reference to the genomic sequence of Non-pathogenic E. coli type K12, Genbank U82664). On the other hand, the 3 'end of this fragment (nucleotides 2457 to 4418) and the entire 387 bp sequence of the plasmid pSP7 do not exhibit any homology with the sequence of non-pathogenic strains of E. coli (Figure 1).

These results demonstrate that the genome of E. coli 0157: H7 contains a locus inserted between positions 93848 and 93849, with reference to the genomic sequence of E. non-pathogenic coli type K12 (Genbank U82664). The results obtained in Example 1 also indicate that this locus is also present in other STECs of serotype 0157 but that it is absent from strains of colibacilli producing shigatoxins of other serotypes as well as strains of E. coli serotype 0157 not producing shiga-toxins.

2. 3. Complete sequencing of the locus inserted in the chromosome of E. coli 0157: H7
The sequence of the locus inserted into the chromosome of E. coli 0157: H7 was completed from the amplification product obtained with the P388D forward primers (position 4205 to 4230 of the 4.4 kb fragment of pSP26, Table I) and P388P antisense (positions 93436 to 93460 of E. coli nonpathogenic type K12, Table I).

The PCR reaction is carried out in a final volume of 100 l containing, 3 μl of total DNA purified as described in example 1, 10 l of 10 times concentrated PCR buffer (100 mM Tris HC1 pH 8.5, 250 mM KCl, 50 mM (NH4) 2 SO4, 20 mM MgSO4), 200 M of each deoxynucleotide, 600 nM of each primer and 2.5 IU of PWO DNA polymerase (ROCHE MOLECULAR).

PCR amplification is carried out under the following conditions: an initial step of denaturation at 94 C for 2 min is followed by 30 cycles alternating a step of denaturation at 94 C for 40 s, a step of hybridization of the primers at 69 C, for 40 s and an extension step at 72 C,

<Desc / Clms Page number 23>

for 50 s and a final elongation step is carried out at 72 C for 10 min.

These primers do not amplify any E. non-pathogenic K12 type coli, on the other hand a 1299 bp amplification product called Amp 388P / 388D, (FIG. 1) is amplified from the DNA of E. coli 0157: H7 (strain EDL 933).

The relative arrangement of the different sequenced fragments is illustrated in Figure 1.

The sequence of the 1299 bp amplification product shows that the 413 bp of its 3 'end has 92% identity with the sequence of E. non-pathogenic coli (positions 93848 to 93436, with reference to the genome sequence of non-pathogenic E. coli type K12, Genbank U82664).

The sequences of the 1299 bp amplification product and of the 4.4 kb fragment of pSP26 made it possible to identify a new sequence of 5478 bp (SEQ ID NO: 1), in E. coli 0157: H7. This sequence contains a small 2634 bp locus (positions 2457 to 5090 of the 5478 bp sequence) inserted into the chromosome of E. coli 0157: H7, flanked on both sides by the conserved sequences of E. coli 0157: H7. non-pathogenic coli. This locus called SILO, 57 (Small Inserted Locus in STEC 0157) is presented in the list in the appendix under the number SEQ ID NO: 2.

The sequence of SILO157 shows no homology with a known sequence and has a G + C content of 44.4%, which is markedly lower than that of the entire E. coli (50.8%, BLATTNER et al., 1997, Science, 277: 1453-1474). This locus is inserted between positions 93848 and 93849, with reference to the sequence of the non-pathogenic E.coli type K12 genome, Genbank U82664).

Remarkably, in enterohemorrhagic colibacilli of serotype 0157: H7, this locus is precisely inserted between the stop codons of 2 open reading frames (Genbank AAB40241, homologous to the open reading frame HI0293 of Haemophilus influenzae and Genbank AAB40242) which are in opposite orientation. In contrast, in the sequence of E. non-pathogenic K12 type coli, the stop codons of these 2 open reading frames are directly joined.

<Desc / Clms Page number 24>

3. Analysis of the potential open reading frames of the SILO157 locus
The SILO locus, .57 comprises 4 potential open reading frames of more than 100 amino acids, called respectively IHP1 (Inserted Hypothetical Protein 1), IHP2 (Inserted Hypothetical Protein 2), IHP3 (Inserted Hypothetical Protein 3) and IHP4 (Inserted Hypothetical Protein 3) Protein 4), corresponding respectively to proteins of 338 (SEQ ID NO: 18), 140 (SEQ ID NO: 20), 120 (SEQ ID NO: 22) and 159 (SEQ ID NO: 24) amino acids. The positive strand encodes IHP1 (nucleotides from positions 3364 to 4380 of the sequence of 5478 bp, SEQ ID NO: 17) and IHP2 (nucleotides from positions 4465 to 4887 of the sequence of 5478 bp SEQ ID NO: 19), in the same phase, and the negative strand codes IHP3 (nucleotides of positions 2997 to 2635 of the sequence of 5478 bp SEQ ID NO: 21) and IHP4 (nucleotides of positions 4002 to 3523 of the sequence of 5478 bp SEQ ID NO: 23), in the same phase.

Analysis of the IHP1 protein, using the PSORT and BLAST programs respectively indicates the presence of a cleavable N-terminal sequence of 23 amino acids, and homologies with membrane proteins and bacterial adhesins. In particular, one can note homologies with the following proteins; a potential invasive / adhesin from Neisseria meningitidis (Genbank AAF 42321 and Genbank AAF41395), Yopl invasin from Yersinia pseudotuberculosis (Genbank CAA32088), YadA adhesin from Yersinia enterocolitica (Genbank CAA32086), HIA adhesin (Genbank AAC437) HSF protein (product of the fibril locus, referred to as hsf locus, Genbank accession number AAC44560) from Haemophilus influenzae, Salmonella rubislaw flagellin E-1 (Genbank CAA28190) and autolysin / adhesin of the surface protein Aas of Staphylococcus saprophyticus (Genbank CAA03852).

Analysis of the IHP2 protein indicates the presence of an N-terminal cleavable sequence of 27 amino acids and little homology with known proteins, except with the YCCF protein of Bacillus subtilis (Genbank CAB12066).

<Desc / Clms Page number 25>

IHP3 shows homologies with a human renal phosphate-dependent sodium transporter (Genbank AAA36354), as well as with the transposase of Bacillus thuringiensis (Genbank Q99335).

IHP4 does not show homology with known proteins.

All of these results show that the enterohemorrhagic E. coli strain 0157: H7 contains a small locus (SILO, 57) inserted into its chromosome. The insertion of this locus is specific to strains of E. coli producing shigatoxins of serotype 0157 (STEC-0157) and comprises 4 potential open reading frames, one of which (IHP1) probably codes for an adhesin-type surface membrane protein which could play a role in the pathogenicity of STEC of serotype 0157.

EXAMPLE 2: DETECTION OF STEC-0157 BY PCR 1. DNA extraction and PCR reaction
211 strains of E. coli. 56 STEC strains of serotype 0157,93 STEC strains of other serotypes, 55 strains of E. coli non-producing shiga-toxins and 7 strains of bacteria of other species, from reference laboratories or isolated from food or biological samples, are cultured in TSB (Tryptone Soy Broth) medium, at 37 C, overnight. Total bacterial DNA is extracted using the INSTAGENE Matrix (BIO-RAD), following the manufacturer's instructions.

The PCR is carried out using the GENEAMP PCR SYSTEM 9600 (PERKIN ELMER) and primers obtained by conventional techniques of oligonucleotide synthesis. The sequence of the various primers used is presented in Table I below, in which the positions in SILO, 57 relate to the sequence of the fragment of 5478 bp (SEQ ID NO: 1).

<Desc / Clms Page number 26>

TABLE 1

<tb>
<tb> Position
<tb> Primer <SEP> Position <SEP> in <SEP> in <SEP> E. <SEP> coli <SEP> sequence <SEP> number
<tb> SILO157 <SEP> K12
<tb> P388C <SEP> 4504-4479 <SEP> absent <SEP>5'-CCGCCAGCAGGGAAAACGATTTAATG-3'<SEP>'<SEP> SEQ <SEP> ID <SEP> NO <SEP> 8
<tb> P388D <SEP> 4205-4230 <SEP> 1 <SEP> absent <SEP>5'-TTAAAACCGGTGACGTGATGATGGTG-3'<SEP> SEQ <SEP> ID <SEP> NO.6
<tb> P388F <SEP> 4329-4305 <SEP> absent <SEP>5'-CGCAGAAATACCGGCTTTAAGTACC-3'<SEP> SEQ <SEP> ID <SEP> NO <SEP> 7
<tb> P388G <SEP> 3752-3776 <SEP> absent <SEP>5'-AGCAACAGGCGCAGATCGTAGCCAC-3'<SEP> SEQ <SEP> ID <SEP> NO <SEP>: <SEP> 5
<tb>

The following pairs of primers: P388C / P388D, P 388F / P388D and P388F / P388G are used in a reaction volume of 50 l containing 200 M of each of the deoxynucleotides (dNTP), 600 nM of each of the primers, 5 l of buffer 10X GENE AMP (PERKIN ELMER), 2.5 IU of AMPLI TAQ GOLD (PERKIN ELMER) and 3 L (approximately 10 ng) of purified bacterial DNA. PCR amplification is carried out under the following conditions: an initial step of denaturation at 94 C for 10 min is followed by 35 cycles alternating a step of denaturation at 94 C for 25 s, a step of hybridization of the primers at 71 C, for 30 s and an extension step at 72 C, for 30 s and a final elongation step is carried out at 72 C for 10 min. The amplification products are separated by electrophoresis in 2% agarose gel containing ethidium bromide, in TBE buffer (Tris-Borate-EDTA) and visualized under an ultraviolet lamp. The 1 kb scale (BIOLABS NEW ENGLAND) or the molecular weight marker VI (ROCHE MOLECULAR) are used as a size marker.

The size of the fragments expected by amplification of a STEC of serotype 0157: H7 (FIG. 1) or of a strain of E. non-pathogenic K12 type coli, with the different pairs of primers, is presented in Table II below.

TABLE II

<tb>
<tb> Size <SEP> of the <SEP> amplification product <SEP><SEP> (bp)
<tb> Pair <SEP> of primers <SEP> STEC <SEP> O157 <SEP>: H7 <SEP> E. <SEP> coli <SEP> K12
<tb> P388C / P388D <SEP> 300 <SEP> (SEQ <SEP> ID <SEP> NO <SEP> 12) <SEP> none
<tb> P388F / P388D <SEP> 125 <SEP> (SEQ <SEP> ID <SEP> NO <SEP> 13) <SEP> none
<tb> P388F / P388G <SEP> 578 <SEP> (SEQ <SEP> ID <SEP> NO <SEP> 14) <SEP> none
<tb>

<Desc / Clms Page number 27>

2. Study of the specificity of the pairs of primers P388C / P388D, P388F / P388D and P388F / P388G a) pair of primers P388C / P388D
In a first series of experiments, the pair P388C / P388D was used under the amplification conditions described above. The results obtained with the 211 strains tested are as follows: - a product of 300 bp is amplified in all the STEC strains of serotype 0157, no product is amplified in the majority of the strains of colibacilli producing shiga-toxins from other serotypes (STEC-non 0157) - no product is amplified in E. coli serotype 0157 not producing shiga-toxins (0157-non STEC), - no product is amplified in bacteria of other species, in particular Hafnia alvei and Citrobacter.

The results obtained show that the pair of primers used makes it possible to specifically detect the STEC of serotype 0157. In addition, the use of this detection method avoids the problems of cross-reactions of the anti-0157 sera with the antigens of Hafnia alvei. or Citrobacter, observed with immunological methods (ELISA).

It was also observed that when the hybridization temperature of the primer pair P388C / P388D was lowered to 60 ° C., an amplification product was observed in the strains of E. coli of the following serotype (055, 077,079, 145 and 0127: H6). b) pairs of primers P388F / P388D and P388F / P388G
To increase the specificity of the detection of STEC serotype 0157, two other primers were tested: the primer P388F whose position is internal to the 300 bp fragment amplified using the primers P388C / P388D, and the primer P388G , located 5 'of this sequence (Table I and Figure 1). These primers were selected so that the

<Desc / Clms Page number 28>

P388F, P388G and P388D primers which are used in pairs (P388F / P388D and P388F / P388G) have similar melting temperatures.

The same 211 strains of bacteria were tested with the 2 pairs of primers P388F / P388D and P388F / P388G and the results obtained with these strains are illustrated in the first 2 columns of Tables III, IV and V below.

TABLE III

<tb>
<tb> Strains <SEP> Number
<tb> from E. <SEP> coli <SEP> of <SEP> Number <SEP> of <SEP> strains <SEP> positive <SEP> in <SEP> PCR / number <SEP> of <SEP> strains <SEP> tested
<tb> strains
<tb> STEC <SEP> Total: 149 <SEP> P388F / P388G <SEP> P388F / P388D <SEP> P388P / P388L <SEP> P388UP388M <SEP> P388P / P388
<tb> (elongation <SEP> 40s)
<tb> 0157.H7 <SEP> 34 <SEP> 34/34 <SEP> 34/34 <SEP> 0/34 <SEP> 34/34 <SEP> 34/34
<tb> 0157 / H7 (P) a <SEP> 4 <SEP> 4/4 <SEP> 4/4 <SEP> 0/4 <SEP> 4/4 <SEP> 4/4
<tb> 0157. <SEP> NM <SEP> 5 <SEP> 5/5 <SEP> 5/5 <SEP> 0/5 <SEP> 5/5 <SEP> 5/5
<tb> 0157.H- <SEP> 1 <SEP>! <SEP> 1/1 <SEP> 1/1 <SEP> 0/1 <SEP> 1/1 <SEP> 1/1
<tb> 0157.ND <SEP> 12 <SEP> 12/12 <SEP> 12/12 <SEP> 0/12 <SEP> 12/12 <SEP> 12/12
<tb> 03 <SEP> 2 <SEP> 0/2 <SEP> 0/2 <SEP> 2/2 <SEP> 0/2 <SEP> 0/2
<tb> 04-NM <SEP> 1 <SEP> 0/1 <SEP> 0/1 <SEP> 1/1 <SEP> 0/1 <SEP> 0/1
<tb> 05 <SEP> 2 <SEP> 0/2 <SEP> 0/2 <SEP> 2/2 <SEP> 0/2 <SEP> 0/2
<tb> 06 <SEP> 9 <SEP> 0/9 <SEP> 0/9 <SEP> 9/9 <SEP> 0/9 <SEP> 0/9
<tb> 08: H19 <SEP>, <SEP> 1 <SEP> 0/1 <SEP> 0/1 <SEP> 1/1 <SEP> 0/1 <SEP> 0/1
<tb> 022 <SEP> 1 <SEP> 0/1 <SEP> 0/1 <SEP> 1/1 <SEP> 0/1 <SEP> 0/1
<tb> 026 <SEP> 14 <SEP> 1 <SEP> 0/14 <SEP> 0/14 <SEP> 14/14 <SEP> 0/14 <SEP> 0/14
<tb> 045: H2 <SEP> 1 <SEP> 0/1 <SEP> 0/1 <SEP> 1/1 <SEP> 0/1 <SEP> 0/1
<tb> 053 <SEP> 2 <SEP> 0/2 <SEP> 0/2 <SEP> 2/2 <SEP> 0/2 <SEP> 0/2
<tb> 055: H7 <SEP> 1 <SEP> 1/1 <SEP> 1/1 <SEP> 0/1 <SEP> 1/1 <SEP> 1/1
<tb> 076 <SEP> 1 <SEP> 0/1 <SEP> 0/1 <SEP> 1/1 <SEP> 0/1 <SEP> 0/1
<tb> 077 <SEP> 3 <SEP> 0/3 <SEP> 0/3 <SEP> 0/3 <SEP> 1/3 <SEP> 1/3
<tb> 079 <SEP> 1 <SEP> 0/1 <SEP> 0/1 <SEP> 0/1 <SEP> 1/1 <SEP> 1/1
<tb> 088: ND <SEP> 1 <SEP> 0/1 <SEP> 0/1 <SEP> 1/1 <SEP> 0/1 <SEP> 0/1
<tb> 091 <SEP> 5 <SEP> 5 <SEP> 0/5 <SEP> 0/5 <SEP> 5/5 <SEP> 0/5 <SEP> 0/5
<tb> 0103: H2 <SEP> 7 <SEP> 0/7 <SEP> 0/7 <SEP> 7/7 <SEP> 0/7 <SEP> 0/7
<tb> 0110 <SEP> 2 <SEP> 0/2 <SEP> 0/2 <SEP> 2/2 <SEP> 0/2 <SEP> 0/2
<tb> 0111 <SEP>, <SEP> 13 <SEP> 0/13 <SEP> 0/13 <SEP> 13/13 <SEP> 0/13 <SEP> 0/13
<tb> 0113 <SEP> 3 <SEP> 0/3 <SEP> 0/3 <SEP> 3/3 <SEP> 0/3 <SEP> 0/3
<tb> 0117 <SEP> 2 <SEP> 0/2 <SEP> 0/2 <SEP> 2/2 <SEP> 0/2 <SEP> 0/2
<tb> 0118: H <SEP> 1 <SEP> 0/1 <SEP> 0/1 <SEP> 1/1 <SEP> 0/1 <SEP> 0/1
<tb> 0125 <SEP> 1 <SEP> 0/1 <SEP> 0/1 <SEP> 1/1 <SEP> 0/1 <SEP> 0/1
<tb> 0126: H8 <SEP> 1 <SEP> 0/1 <SEP> 0/1 <SEP> 1/1 <SEP> 0/1 <SEP> 0/1
<tb> 0128: H2 <SEP> 1 <SEP> 0/1 <SEP> 0/1 <SEP> 1/1 <SEP> 0/1 <SEP> 0/1
<tb> 0136 <SEP> 2 <SEP> 0/2 <SEP> 0/2 <SEP> 2/2 <SEP> 0/2 <SEP> 0/2
<tb> 0139-ND <SEP> 1 <SEP> 0/1 <SEP> 0/1 <SEP> 1/1 <SEP> 0/1 <SEP> 0/1
<tb> 0141ac <SEP> 1 <SEP> 0/1 <SEP> 0/1 <SEP> 1/1 <SEP> 0/1 <SEP> 0/1
<tb> 0145 <SEP> 3 <SEP> 0/3 <SEP> 0/3 <SEP> 0/3 <SEP> 0/3 <SEP> 0/3
<tb> 0147 <SEP> 1 <SEP> 0/1 <SEP> 0/1 <SEP> 1/1 <SEP> 0/1 <SEP> 0/1
<tb> NT <SEP>, <SEP> 9 <SEP> 0/9 <SEP> 0/9 <SEP> 9/9 <SEP> 0/9 <SEP> 0/9
<tb>
cured strains of the OU MUa plasmid

<Desc / Clms Page number 29>

TABLE IV:

<tb>
<tb> Strains <SEP> Number <SEP> Number <SEP> of <SEP> strains <SEP> positive <SEP> in <SEP> PCR / number <SEP> of <SEP> strains <SEP> tested
<tb> from E. <SEP> coli <SEP> from
<tb> strains
<tb> non-STEC <SEP> Total <SEP>: 55 <SEP> P388F / P388G <SEP> P388F / P388D <SEP> P388P / P388L <SEP> P388UP388M <SEP> P388P / P388D
<tb> (elongation <SEP> 40s)
<tb> 0157-H- <SEP> 2 <SEP> 1 <SEP> 0/2 <SEP> 0/2 <SEP> 2/2 <SEP> 0/2 <SEP> 0/2
<tb> 0157: ND <SEP> 40 <SEP> 0/40 <SEP> 0/40 <SEP> 40/40 <SEP> 0/40 <SEP> 0/40
<tb> 0157: H19 <SEP> 2 <SEP> 0/2 <SEP> 0/2 <SEP> 2/2 <SEP> 0/2 <SEP> 0/2
<tb> 026 <SEP> 3a <SEP> 0/3 <SEP> 0/3 <SEP> 3/3 <SEP> 0/3 <SEP> 0/3
<tb> 055 <SEP> 3b <SEP> 3/3 <SEP> 3/3 <SEP> 0/3 <SEP> 3/3 <SEP> 3/3
<tb> 0103 <SEP> 1 <SEP> 0/1 <SEP> 0/1 <SEP> 1/1 <SEP> 0/1 <SEP> 0/1
<tb> 0111 <SEP> 2c <SEP> 0/2 <SEP> 0/2 <SEP> 2/2 <SEP> 0/2 <SEP> 0/2
<tb> 0127: H6 <SEP> 1d <SEP> 1/1 <SEP> 0/1 <SEP> 0/1 <SEP> 0/1 <SEP> 0/1
<tb>

(E2348/69)~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

(E2348 / 69) ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

<tb>
<tb> 0128: H2 <SEP> 1e <SEP> 0/1 <SEP> 0/1 <SEP> 1/1 <SEP> 0/1 <SEP> 0/1
<tb>
a: 2 out of 3 strains are EPEC b: all 3 strains are EPEC c: 1 in 2 strain is a strain of E. colientoaggregative d: This strain is the reference strain for EPEC e: this strain is an EPEC TABLE V:

<tb>
<tb> Others <SEP> Number <SEP> Number <SEP> of <SEP> strains <SEP> positive <SEP> in <SEP> PCR / number <SEP> of <SEP> strains <SEP> tested
<tb> bacteria <SEP> from
<tb> strains
<tb>, <SEP> 1 <SEP> Total: 7 <SEP> P388F / P388G <SEP> P388F / P388D <SEP> P388P / P388L <SEP> P388L / P388M <SEP> P388P / P388D
<tb> (elongation <SEP> 40s)
<tb> Enterobacter <SEP> 1 <SEP> 0/1 <SEP> 0/1 <SEP> 0/1 <SEP> 0/1 <SEP> 0/1
<tb> Citrobacter <SEP> 1 <SEP> 0/1 <SEP> 0/1 <SEP> 0/1 <SEP> 0/1 <SEP> 0/1
<tb> Hafnia <SEP> alvei <SEP>, <SEP> 5 <SEP> 0/5 <SEP> 0/5 <SEP> 0/5 <SEP> 0/5 <SEP> 0/5
<tb>

These results show that: - all the strains of colibacilli producing shiga-toxins of serotype 0157 (56 strains of STEC of serotype 0157 tested including 38 strains of serotype 0157: H7 including 4 strains cured of the 60 Mda plasmid, 1 strain of serotype 0157: H- and 17 other STEC strains of serotype 0157 (5 0157: NM (Not Mobile) and 12 0157: ND (Not Determined)] are amplified by the pairs of primers P388F / P388D and P388F / P388G. , the observation of an amplification product in the strains of E. coli 0157: H7 cured of the 60 MDa plasmid also confirms the chromosomal location of the SIL0157 locus.

- the strains of colibacilli producing shiga-toxins of other serotypes (30 different serotypes tested on a total of 93 strains) are not amplified with the pairs of primers P388F / P388D and P388F / P388G. In particular serotypes 0145,077 and 079 which are amplified

<Desc / Clms Page number 30>

with the primers P388C / P388D are not amplified with these two pairs of primers.

- none of the serotype 0157 strains which do not produce shiga-toxins (44 strains tested) is amplified with the pairs of primers P388F / P388D and P388F / P388G,
7 strains of 3 other bacterial genera including those which give cross reactions with 0157 by immunological techniques (Citrobacter and Hafnia alvei) are not amplified with the pairs of primers P388F / P388D and P388F / P388G, the strains of E. coli serotype 055 (STEC and non-STEC) are also amplified with the pairs of primers P388F / P388D and P388F / P388G, which is explained by the clonal origin of serotypes 055 and 0157 which has been clearly demonstrated (FENG et al., 1998, J. Infect. Dis., 177: 1750-1753; WHITTAM et al., 1993, Infect. Immun., 61: 1619-1629). strains of E. coli non-producing shiga-toxins of other serotypes (non-STEC and non-0157; serotypes tested on a total of 8 strains) are not amplified by the pairs of primers P388F / P388D and P388F / P388G, at the exception of 0127: H6 which is amplified with the couple P388F / P388G but not with the couple P388F / P388D.

All the results demonstrate that the two pairs of primers P388F / P388G and P388F / P388D make it possible to detect all the strains of colibacilli producing shiga-toxins of serotype 0157. The pair P388F / P388D however has a higher specificity. than the couple P388F / P388G.

<Desc / Clms Page number 31>

EXAMPLE 3: ANALYSIS OF THE CONSERVATION OF THE SITE OF INSERTION OF LOCUS SILO, 57 IN E. COLI AND DEMONSTRATION OF A HOMOLOGATED LOCUS IN THE GROUP OF ENTEROPATHOGENIC COLIBACILLI (EPEC) 1. Analysis of the conservation of the insertion site of the SILO locus, 57 in E. coli
The conservation of the insertion site of the SILO locus, 57 was studied by PCR in the 211 strains of bacteria analyzed in Example 2, using the following 3 pairs of primers: the pairs P388L / P388M and P388P / P388D which respectively amplify the left and right junctions of the SILO157 locus and the P388P / P388L pair which amplifies a product of 682 bp in all the strains of E. coli non-pathogenic such as the K12 strain (Figure 1).

The reactions are prepared according to the protocol described in Example 1 and the PCR amplification is carried out under the following conditions: an initial denaturation step at 94 C for 10 min is followed by 35 cycles alternating with a denaturation step at 94 C for 40 s, a step of hybridization of the primers at 65 C, for 40 s and an extension step at 72 C, for 50 s and a final elongation step is carried out at 72 C for 10 min.

The sequences of the primers used and the size of the fragments expected with the STEC-0157 or with the E. coli K12 strain are presented respectively in Tables VI and VII below. In Table VI, the positions in SILO, 57 relate to the 5478 bp fragment (SEQ ID NO: 1) and the positions in E.coli K12 relate to the Genbank U82664 sequence.

TABLE VI:: ~~~~~~~~~~~~~~~~~~~~~~

<tb>
<tb> Start <SEP> Position <SEP> Position <SEP> number
<tb> in <SEP> in <SEP> E. <SEP> sequence
<tb> SILO157 <SEP> coli <SEP> K12 <SEP>
<tb> P388M <SEP> 2606-2581 <SEP> absent <SEP>5'-GGCGGAGGCGGCCTACGGTTTTGGCG-3'<SEP> SEQ <SEP> ID <SEP> NO <SEP> 4
<tb> P388P <SEP> 5503-5479 <SEP> 93436-93460 <SEP>5'-GAACGATGAACATCAGCGATGTAGC-3'<SEP> SEQ <SEP> ID <SEP> NO.9
<tb>

P388L 2188-2213 94117-94092 5'-CGCTGCTCGCCGCCTGGCTGTGGTGG-3' 1 SEQ ID NO:3

P388L 2188-2213 94117-94092 5'-CGCTGCTCGCCGCCTGGCTGTGGTGG-3 '1 SEQ ID NO: 3

<Desc / Clms Page number 32>

TABLE VII

<tb>
<tb> Size <SEP> of the <SEP> amplification product <SEP><SEP> (bp)
<tb> Pair <SEP> of primers <SEP> STEC <SEP> 0157: <SEP> H7 <SEP> E. <SEP> coli <SEP> K12
<tb> P388P / P388D <SEP> 1299 <SEP> (SEQ <SEP> ID <SEP> NO <SEP>: 15) <SEP> none
<tb> P388P / P388L <SEP> 3316 <SEP> (SEQ <SEP> ID <SEP> NO <SEP>: 11) <SEP> 682 <SEP> (SEQ <SEP> ID <SEP> NO <SEP>: 10)
<tb> P388L / P388M <SEP> 419 <SEP> (SEQ <SEP> ID <SEP> NO <SEP>: 16) <SEP> none
<tb>

The results are illustrated in the last 3 columns of Tables III, IV and V.

The results obtained show that by using short elongation times (40s) for the pair P388P / P388L, no amplification product is observed in all the strains of colibacilli producing shigatoxins of serotype 0157, and that the majority other strains are identical to the strain of E. coli K12 in the 9-11 minute region and does not contain the SILO locus, 57-
However, some strains tested have a particular amplification profile:
1) The strains of serotype 055 (STEC and nonSTEC), the strain of serotype 079 (STEC) and a strain of serotype 077 (STEC) have a profile similar to that of STEC of serotype 0157, which suggests the presence, in these strains of a locus homologous to SILO, 57, inserted between positions 93849 and 93848 of the genome of E. coli., with reference to the genome of non-pathogenic E. coli type K12 (Genbank U82664)
However, with the exception of the E. coli strains of serotype 055, which, due to their relationship to serotype 0157, are amplified by the primers P388F / P388D and P388F / P388G, the locus homologous to SIL0157, present in the colibacilli of serotype 079 and 077 is not amplified by primers P388F / P388D and P388F / P388G. Consequently, the presence of a locus homologous to SILO, .57 in these colibacilli does not prevent the specific detection of STEC of serotype 0157 using the primers P388F / P388D and P388F / P388G.

2) 2 strains of serotype 077 (STEC), 3 strains of serotype 0145 (STEC) and one strain of EPEC (0127: H6) give negative results with the 3 pairs of primers, which

<Desc / Clms Page number 33>

indicate that they have, inserted in this same region, a locus whose sequence differs significantly from that of SILO157-2. Identification of a locus homologous to the SILO157 locus in an EPEC (0127: H6)
The EPEC strain E2348 / 69 (0127: H6) was amplified using the pair of primers P388P / P388L, under the following conditions which make it possible to amplify large products: an initial stage of denaturation at 94 C for 10 min is followed by 35 cycles alternating a denaturation step at 94 C for 25 s, a primer hybridization step at 62 C, for 25 s and an extension step at 72 C, for 2 min 15 s and a final elongation step is carried out at 72 ° C. for 7 min. The strain of E. coli 0157: H7 (strain EDL 933) is used as a positive control for amplification.

The products obtained were hybridized with a probe called probe FD, corresponding to the amplification product of the strain 0157: H7 using the primers P388F and P388D, under the conditions described in Example 1.

The results obtained are as follows: an amplification product of 682 bp which does not hybridize with the FD probe is observed with the strains which do not have a locus inserted between minutes 9 and 11 of the E. coli non-pathogenic, - an amplification product of 3316 bp, hybridizing with the FD probe is observed with the strain 0157: H7 and - an amplification product of size slightly greater than 3316 bp, hybridizing with the FD probe is observed with strain 0127: H6.

These results demonstrate that the EPEC 0127: H6 strain contains a locus homologous to the SILO157 locus.

They also demonstrate that the SILO157 locus isolated from the EDL933 strain of STEC-0157: H7 is present in:

<Desc / Clms Page number 34>

- all colibacilli producing shigatoxins of serotype 0157, - pathogenic E. coli strains of serotype 055 (STEC or EPEC); which is explained by the clonal origin of serotypes 055 and 0157, and - in other serotypes of STEC, in particular in the colibacilli producing shiga-toxins of serotype 077,079 and 0145.

The presence of the SILO157 locus or of a homologous locus inserted between positions 93848 and 93849 of the E.coli genome (with reference to the non-pathogenic E.coli genome of type K12, Genbank U82664) therefore constitutes a new means of specifically detect pathogenic E. coli strains and in particular all STEC-0157.

Claims (29)

1) Nucleic acid molecule derived from a pathogenic strain of E. coli, characterized in that it is selected from the group consisting of: a) a nucleic acid molecule representing a locus inserted into the genome of said pathogenic strain, at the location of said genome corresponding to that located in E.coli K12, between positions 93848 and 93849 of the GenBank sequence U82664; b) a nucleic acid molecule comprising the molecule a) defined above, and between 1 and 2500 bp of the sequence upstream of the site of insertion of said molecule, and / or between 1 and 2500 bp downstream of the site insertion of said molecule. c) a nucleic acid molecule constituting a fragment of at least 15 bp, preferably at least 25 to 30 bp, of the molecule a) above; d) a nucleic acid molecule constituting a fragment of at least 15 bp, preferably at least 25 to 30 bp, of the above molecule b), said fragment comprising at least a portion of the molecule a) and at least a portion of the sequence upstream of the insertion site of said molecule a), or at least a portion of the sequence downstream of the insertion site of said molecule a). 2) Nucleic acid molecule according to claim 1, characterized in that said pathogenic bacterium is chosen from Escherichia coli STEC of serotype 0157,077, 079, or 0145, Escherichia coli STEC or EPEC of serotype 055, and Escherichia coli EPEC serotype 0127. 3) Nucleic acid molecule according to any one of claims 1 or 2, characterized in that it is chosen from the sequences: SEQ ID NO: 1, SEQ ID NO: 2, SEQ ID NO: 11 to SEQ ID NO: 17, SEQ ID NO: 19, SEQ ID NO: 21 and SEQ ID NO: 23. 4) A nucleic acid molecule capable of hybridizing specifically under stringent conditions with a <Desc / Clms Page number 36> A nucleic acid molecule according to any one of claims 1 to 3. 5) A nucleic acid molecule according to claim 4, chosen from oligonucleotides P388L (SEQ ID NO: 3), P388M (SEQ ID NO: 4), P388G (SEQ ID NO: 5), P388D (SEQ ID NO 6) , P388F (SEQ ID NO: 7), P388C (SEQ ID NO: 8) and P388P (SEQ ID NO: 9). 6) Use of a nucleic acid molecule according to any one of claims 1 to 5 for the screening of pathogenic E. coli. 7) Screening process for E. pathogenic coli, characterized in that it comprises the detection of the presence in the genome of E. coli bacteria present in the test sample, of a locus inserted at the location of said genome corresponding to that located in E. coli K12, between positions 93848 and 93849 of the GenBank sequence U82664. 8) Method according to claim 7, characterized in that it comprises at least one step of polymerase chain reaction using a pair of primers comprising a primer hybridizing with the region of the genome of E.coli defined by the sequence 2500 bp upstream from position 93848 of the GenBank U82664 sequence, and a primer hybridizing with the region of the E.coli genome defined by the sequence 2500 bp downstream from position 93849 of the GenBank sequence U82664. 9) Method according to claim 8, characterized in that one uses the pair of primers P388L / P388P. 10) Method according to claim 7, characterized in that it comprises at least one step of polymerase chain reaction using a pair of primers hybridizing with internal sequences of a molecule a) according to claim 1. 11) Method according to claim 10, characterized in that one uses a pair of primers chosen from the use of the following pairs of primers: P388C / P388D, P388F / P388D, P388F / P388G. <Desc / Clms Page number 37> 12) The method of claim 7, characterized in that it comprises at least one step of polymerase chain reaction using a pair of primers comprising a primer hybridizing with an internal sequence of a molecule a) as defined above, and a primer hybridizing with the region of the E.coli genome defined by the sequence of 2500 bp upstream of position 93848 of the sequence GenBank U82664, or a primer s' hybridizing with the region of the E.coli genome defined by the 2500 bp sequence downstream of position 93849 of the GenBank U82664 sequence. 13) Method according to claim 12, characterized in that said pair of primers is chosen from P388P / P388D and P388L / P388M. 14) Method according to any one of claims 7 to 13, characterized in that it is used for the detection of E. coli of serotype 0157 producing shiga-toxins 15) Protein encoded by an open reading frame present on a nucleic acid molecule according to any one of claims 1 to 3, and in particular on a molecule derived from a strain of E. coli of serotype 0157 producing shiga toxins . 16) Protein according to claim 15, characterized in that it is chosen from the proteins IHP1, IHP2, IHP3, and IHP4 of respective sequences SEQ ID NO: 18, 20, 22 and 24. 17) Peptide, characterized in that it consists of a fragment of at least 5 amino acids of a protein according to any one of claims 15 or 16. 18) A nucleic acid molecule according to claim 1, encoding a protein or a peptide according to any one of claims 15 to 17. 19) A nucleic acid molecule according to claim 18 characterized in that it comprises a sequence selected from the group consisting of the sequences SEQ ID NO: 17,19, 21 and 23. 20) Expression cassette, characterized in that it comprises a nucleic acid molecule encoding a <Desc / Clms Page number 38> protein according to any one of claims 18 or 19 placed under the transcriptional control of an appropriate promoter. 21) Nucleic acid vector, characterized in that it comprises an insert consisting of a nucleic acid molecule according to any one of claims 1 to 4, 18, or 19, or an expression cassette according to claim 20. 22) Cell transformed by at least one vector according to claim 21. 23) Antibody, directed against a protein or a peptide according to any one of claims 15 to 17. 24) Use of a protein or a peptide according to any one of claims 15 to 17 for the detection of pathogenic strains of E. coli. 25) Use of an antibody according to claim 23 for the detection of pathogenic strains of E. coli. 26) Use according to any one of claims 6, 24, or 25, characterized in that said pathogenic E. coli strains are strains of E. coli of serotype 0157 producing shiga toxins, 27) Use according to any one of claims 6, or 24 to 26, characterized in that the search for pathogenic E.coli strains is carried out in a food product. 28) Use according to any one of claims 6, or 24 to 26, characterized in that the search for pathogenic E.coli strains is carried out in a faeces sample. 29) Use of a protein or a peptide according to any one of claims 15 to 17 for the preparation of a vaccine.